The germ line enables metazoan organisms to produce gametes and thus is responsible for the fertility and perpetuation of species. We are investigating how the germ line is specified and its early development controlled, taking advantage of the powerful genetics available in the nematode worm C. elegans. The germ cells in many species contain distinctive granules, which may serve as determinants of germ-cell fate. We will elucidate the role of these germ-line "P granules" in C. elegans, by studying two genes that encode presumptive P-granule components. glh-1 was identified on the basis of its similarity to a Drosophila germ-granule component, vasa, and antibodies to GLH-1 stain P granules. pgl-1 was identified in genetic screens: pgl-1 mutant worms lack some P-granule epitopes and show a maternal-effect sterile phenotype (pgl/pgl mothers produce all sterile progeny). We will address the following questions: What is the mutant phenotype of glh-1? If glh-1 mutants are sterile (as predicted by anti- sense experiments), what are the germ-line defects? Is PGL-1 associated with P granules? Does PGL-1 resemble any known proteins? What are the germ-line defects in pgl-1 sterile worms? We will also use proven P- granule components (GLH-1 and perhaps PGL-1) to identify other P-granule components, for genetic and molecular analysis. We have identified four maternal-effect genes, mes-2, mes-3, mes-4, and mes-6, that are required for the normal early development and survival of the germ line. The maternal-effect sterile phenotype displayed by the four mes genes is more severe in XX than XO animals, suggesting that the mes genes participate in some aspect of control of gene expression that is sensitive to chromosome dosage. Consistent with this, MES-2 is similar to Drosophila Enhancer of zeste, which is predicted to control gene expression by regulating higher order chromatin structure. MES-6 is a WD- 40 repeat-containing protein, and MES-3 is a novel protein. We will address the following questions: Are the MES proteins localized to sites consistent with their being regulators of gene expression? Does MESA resemble any known proteins? Do the mes genes give a common mutant phenotype because they control each other's expression or localization, or because the MES proteins interact and function as a complex? Which of three models explains the sensitivity of the mes mutant phenotype to chromosome dosage? One model is that the mes genes control X chromosome dosage compensation in the germ line.